AI strategy for businesses
We take you from scattered AI ideas to a decision basis you can bring to management: which initiatives hold up, what they require and what you should start with. And if you want to go further, we can build it.
Gothenburg
Senior team on site
Defined scope
Fixed scope before we start
Feasible
We build what we propose
Your material
No locked-in method tools
You do not need to know what you want to build. What you do need is a business area you want to improve and a willingness to deprioritize what does not hold up.
Different parts of the organization want different things and every proposal sounds reasonable. You need to know which ones actually hold up, what they require and in what order they make sense.
A prioritized list with value and effort
You can see a concrete area where time goes into manual work, or where decisions rest on too thin a basis — but not whether AI is the right answer or what it would cost to find out.
An assessment of value, risk and feasibility
You need a basis for decision that holds up in a board meeting: what the investment delivers, what it requires of data and systems, what it costs and what happens if you wait.
A decision basis and a plan for a first pilot
Less suitable when you are only looking for a general market overview without a business area to improve.
Case study · Herrljunga municipality
Herrljunga municipality wanted to make the geothermal permit application simpler. We chose a use case where the rules can be checked rather than interpreted, and built a chat-based application flow with automatic validation against the ancient monuments register, the land survey authority and the municipal GIS. The solution is in testing.
An AI strategy is only worth something if it says what you should do first and what is required to make it work. That is the level we deliver at.
01
We go through processes, systems and data flows and talk to the people doing the work. Without that picture it is impossible to say where AI makes a difference — or where it would just get expensive.
02
We identify where AI can save time, raise quality or improve the basis for decisions — and weigh each proposal against data quality, risk, integration needs and effort.
03
You get a plan you can work from: what starts first, who owns what internally, and which data and integration questions must be solved before production.
How an engagement works
We start narrow and agree the scope before we get going. Each phase should give you something you can make a decision on, even if you choose to stop there.
Get a proposed approachWe get to grips with goals, bottlenecks and the decisions that take too long today. We talk to the business, not just to management.
You get: a current-state picture and a list of possible use cases
We weigh each use case against value, risk, data quality and effort — and recommend what you should start with and what you should leave.
You get: prioritization, a decision basis and a cost picture for the implementation
We set up a defined pilot on the first initiative, with ownership and measurement points. If you want us to build it, we can — otherwise you take it forward yourselves.
You get: a pilot plan, ownership split and a plan forward
Once the prioritization is done, these are the engagements that usually become the first real step.
Examples of AI initiatives we have taken from idea to an evaluated solution.
The key things to sort out before you commission a discovery phase — scope, price, delivery and what happens next.
A picture of where you are today, a prioritized list of use cases with assessed value and effort, the data and integration questions that must be solved first, and a plan for a first pilot with ownership and measurement points. The material is made to be taken straight into a management or board meeting.
It is mainly governed by how many business areas are to be analyzed and how quickly we get access to key people. We structure the work in defined steps and agree scope and timeline before we start — you should never have to approve an open-ended bill.
The price depends on the scope — the number of business areas, how many interviews are required and how deep we go into data and system conditions. We define the engagement first and give a fixed price once the scope is clear. Our pricing guide for software development gives a first impression of what the implementation might then cost.
That is up to you. A common next step is a defined pilot on the highest-priority use case. You can run it internally, with another supplier or together with us — the material is written to work either way.
Yes. We are a development company, not a pure analysis house — we build chatbots, RAG solutions, AI features in products and integrations. That is why the recommendations take into account what can actually be built and maintained. But there is no obligation: you can just as easily take the material forward with someone else.
No. A large part of the work is precisely to map what exists, what is missing and which dependencies must be solved before production. The point is that you should know this before you invest, not after.
You do — the analysis, the priorities, the plans and all the underlying material. We do not work in closed method tools or licensed frameworks that make you dependent on us to use the material.
The same people you meet in the first conversation. We are a small team in Gothenburg and do not hand the engagement over to someone else once the contract is signed.
We reply within one business day